Relationships between Loading Rates and Pollutant Removal during Maturation of Gravel‐Bed Constructed Wetlands

Abstract

AbstractRelationships between pollutant mass loading and removal are reported for a series of five pilot‐scale constructed wetlands in their 4th and 5th years of operation. The wetlands received different hydraulic loadings ranging between ∼15 and 70 mm d−1 (9.5‐2 d theoretical retention, respectively). Effluent concentrations of 5‐d carbonaceous biochemical oxygen demand (CBOD), total nitrogen (TN), ammonium N (NH4‐N), and fecal coliforms (FC) broadly followed seasonal patterns in influent wastewater strength. Mean annual mass removals of 58 to 78% suspended solids (SS), 73 to 91% CBOD, 48 to 65% TN, 34 to 60% NH4‐N, 15 to 38% total phosphorus (TP), and 93 to 99.6% of FC were recorded, with removal efficiencies inversely related to loadings. Mass removal rates were monotonically related to loading rates, and could be modeled using a simple plug‐flow, first‐order approach accounting for removal down to nonzero background concentrations. Comparisons with treatment performance recorded for the wetlands soon after commissioning showed relatively constant relationships between mass loading and removal of CBOD, TN, and FC. In contrast, SS and TP removal declined significantly over the same period. Reduced SS removal efficiency appeared to result from clogging of the gravel substratum by refractory organic solids, and reduced TP removal from saturation of substratum sorption capacity and filling of plant storage pools. To improve N removal predictions for wetlands treating ammonium‐rich wastewaters, the use of a combined carbonaceous and nitrogenous BOD term is proposed, which addresses the oxygen dependence of microbial nitrification, the principal rate‐controlling process.